W. G. YanName — Embryology of Eustylochus. 285 



Fig. 14 also suggests that new rays are being formed, the old ones 

 having already disappeared. 



Yet, on the other hand, there are many cases where we can scarcely 

 doubt that the centrosphere simply lengthens and divides, each part 

 carrying its rays with it, although the latter may become reduced in 

 strength. In support of this, it is worth mentioning that although 

 I have a very large amount of good material in this stage, particu- 

 larly of Planocera nebulosa, I have never found a case where the 

 rays had disappeared entirely nor one where I could distinguish an 

 old and a new set of rays with certainty. 



I think therefore that the conclusion is inevitable that the aster 

 rays of the inner pole of the first polar spindle persist in part as rays 

 of the second polar spindle asters, and my preparations further indi- 

 cate that the extent to which they do this, and the extent to which 

 new radiations are developed, varies greatly in individual eggs. 



I do not find the rays at any period penetrating the centrosphere 

 and reaching the centrosome as described by Van der Stricht, Fran- 

 cotte and many who have studied the eggs of animals of other 

 classes, though during the process of forming the spindle the cen- 

 trospheres have a more or less indefinite outline. As the centro- 

 sphere of the first polar spindle elongates, the part of it immediately 

 surrounding the derived centrosomes becomes the centrosphere of 

 #the second polar spindle aster. 



The second polar spindle (Fig. 16) is shorter and wider than the 

 first, the aster rays are fewer and shorter and the chromosomes more 

 or less widely separated in a lateral direction, which is in strong 

 contrast to their crowded position in the late stages of the first polar 

 spindle (Fig. 13). 



The chromosomes of the second polar spindle are many times 

 smaller than in the first and are less easily stained. Their form is 

 usually irregular and they often divide while the spindle is in an 

 early stage (Figs. 14 and 15). 



The cross form described by Klinckowstrom, Francotte and Van 

 der Stricht is also found occasionall}' in my preparations. It is 

 explained by the latter as two short rods, the segments derived from 

 the division of the chromosome lying across each other at right 

 angles. 



Although some specimens appear to support the theory that the 

 division of the chromatin is here a transverse and reducing division, 

 I cannot claim to have determined the relation of this plane of 

 division of the chronfatin to the previous one, as Van der Stricht 



